Internal:CDT3D GPU Smoothing: Difference between revisions

From crtc.cs.odu.edu
Jump to navigation Jump to search
← Older editNewer edit →
Spyridon97 (talk | contribs)
CNF, external, internal
2,385 edits
Spyridon97 (talk | contribs)
CNF, external, internal
2,385 edits
Line 21: Line 21:
## GPU version:
## GPU version:
### Sequential CPU-wise:
### Sequential CPU-wise:
#### Version 1:
#### Version 1 (all points copied):
##### Steps:
##### Steps:
###### Bounding box on GPU, grid(1) block(1)
###### Bounding box on GPU, grid(1) block(1)
###### Original point value on GPU, grid(1) block(1)
###### Original point value on GPU, grid(1) block(1)
###### Grid creation on GPU, grid(1) block(1)
###### Grid creation on GPU, grid(1) block(1)
###### Grid creation on GPU, grid(1) block(1)
###### Grid point values using global values array on GPU, grid(2, 2, 2) block(4, 4, 4)
###### Grid point values using global values array on GPU, grid(2, 2, 2) block(4, 4, 4)
###### Reduction using global values array on GPU, grid(1) block(1)
###### Reduction using global values array on GPU, grid(1) block(1)
 
##### Time total = 18.563 sec
##### Time
###### CPU = 0.79182 sec
###### Steps 1 & 2 = 8.31743 sec
###### Step 3 = 1.03768 sec
###### Step 4 = 6.55488 sec
###### Steo 5 = 1.73327 sec
###### MemCopyH2D = 0.12792 sec
###### MemCopyD2H = 0.00002 sec


### Parallel CPU-wise:
### Parallel CPU-wise:

Revision as of 23:54, 16 March 2021

Presentations

Spring 2021

  1. A Unified Approach for Parallel Mesh Operations by Christos✔
  2. GPU Architecture ✔

Code

Spring 2021

  1. Discrete Mesh Optimization using the paper: Discrete Mesh Optimization on GPU, The results below are using the tet_shape metric from the Verdict library
    1. CPU version:
      1. Sequential:
        1. Dihedral Quality of different smoothing methods:✔
          Vertex_Smoothing_Methods_Dihedral_Angle_Quality.png
        2. Shape Quality of different smoothing methods:✔
          Shape_Quality.png
      2. Parallel:
        1. Dihedral Quality: Sequential vs Parallel:✔
          Discrete_Vertex_Smoothing_Dihedral_Angle_Quality.png
        2. Speed-Up:✔
          Smoothing_Speed-up_ 1_iteration.png
        3. RollBacks✔:
          Percentage_of_Average_Rollbacks_per_Thread.png
    2. GPU version:
      1. Sequential CPU-wise:
        1. Version 1 (all points copied):
          1. Steps:
            1. Bounding box on GPU, grid(1) block(1)
            2. Original point value on GPU, grid(1) block(1)
            3. Grid creation on GPU, grid(1) block(1)
            4. Grid point values using global values array on GPU, grid(2, 2, 2) block(4, 4, 4)
            5. Reduction using global values array on GPU, grid(1) block(1)
          2. Time total = 18.563 sec
            1. CPU = 0.79182 sec
            2. Steps 1 & 2 = 8.31743 sec
            3. Step 3 = 1.03768 sec
            4. Step 4 = 6.55488 sec
            5. Steo 5 = 1.73327 sec
            6. MemCopyH2D = 0.12792 sec
            7. MemCopyD2H = 0.00002 sec
      1. Parallel CPU-wise:
Retrieved from "https://crtc.cs.odu.edu/index.php?title=Internal:CDT3D_GPU_Smoothing&oldid=8354"